Dynamic, Morphotype-Specific Candida albicans β-Glucan Exposure during Infection and Drug Treatment

Morphotype-Specific Candida albicans b-Glucan Exposure during Infection and Drug Treatment. PLoS Pathog 4(12): e1000227. doi:10.1371/journal.ppat.1000227 Dynamic, Morphotype-Specific Candida albicans b- Glucan Exposure during Infection and Drug Treatment Robert T. Wheeler 0 1 Diana Kombe 0 1 Sudeep D. Agarwala 0 1 Gerald R. Fink 0 1 Aaron P. Mitchell, Carnegie Mellon University, United States of America 0 Current address: University of Maine, BMMB Department , Orono, Maine , United States of America 1 Whitehead Institute for Biomedical Research , Cambridge, Massachusetts , United States of America Candida albicans, a clinically important dimorphic fungal pathogen that can evade immune attack by masking its cell wall bglucan from immune recognition, mutes protective host responses mediated by the Dectin-1 b-glucan receptor on innate immune cells. Although the ability of C. albicans to switch between a yeast- or hyphal-form is a key virulence determinant, the role of each morphotype in b-glucan masking during infection and treatment has not been addressed. Here, we show that during infection of mice, the C. albicans b-glucan is masked initially but becomes exposed later in several organs. At all measured stages of infection, there is no difference in b-glucan exposure between yeast-form and hyphal cells. We have previously shown that sub-inhibitory doses of the anti-fungal drug caspofungin can expose b-glucan in vitro, suggesting that the drug may enhance immune activity during therapy. This report shows that caspofungin also mediates b-glucan unmasking in vivo. Surprisingly, caspofungin preferentially unmasks filamentous cells, as opposed to yeast form cells, both in vivo and in vitro. The fungicidal activity of caspofungin in vitro is also filament-biased, as corroborated using yeast-locked and hyphal-locked mutants. The uncloaking of filaments is not a general effect of anti-fungal drugs, as another anti-fungal agent does not have this effect. These results highlight the advantage of studying host-pathogen interaction in vivo and suggest new avenues for drug development. - Funding: This work was supported in part by an investigator-initiated grant from Merck, Inc. to GRF and RTW, the Bushrod H. Campbell and Adah F. Hall Charity Fund grant of the Charles A. King Foundation to RTW, and grant GM40266 from the National Institutes of Health to GRF. Merck, Inc. had no role beyond funding the work. Competing Interests: This work was funded in part by a grant from Merck, Inc. to GRF and RTW. This sponsor had no role in the study or manuscript beyond funding the work. Candida albicans, a common commensal organism of humans, has emerged as an important fungal pathogen clinically due to the immunocompromised status of many patients as well as the ineffectiveness of current anti-fungal drugs [1]. The immune system has many pathways for recognizing and responding to C. albicans, including innate immune Toll-like receptors, lectin receptors, antibody, complement, and mannose binding lectin [2]. A major innate immune receptor for C. albicans is Dectin-1, a lectin that can recognize b-glucan, a unique component of the fungal cell wall. In vitro analysis shows that Dectin-1 recognizes fungal b-glucan (comprising glucan polymers with mixed b1,3and b1,6- linkages), and signals through unique pathways to induce phagocytosis, up-regulation of immune killing mechanisms, and production of pro-inflammatory cytokines [3]. The mechanism of b-glucan signaling through Dectin-1 is unclear because this signature molecule on Candida and other fungi is enveloped by a cell wall mannoprotein layer that masks almost all of the b-glucan from immune recognition and mutes the host immune response [4,5,6,7,8]. On one hand, the b-glucan masking observed in vitro runs contrary to an active role for Dectin-1 recognition of b-glucan during infection [8,9,10]. On the other hand, in vitro Dectin-1 is clearly able to mediate protective responses to fungi and zymosan, a treated fungal particle with exposed b-glucan [9,11,12,13,14]. In addition, work in knock-out mice suggests at least a conditional requirement for Dectin-1 in resistance to the SC5314 strain of C. albicans, which may be dependent on mouse strain and/or C. albicans strain [13,14]. An additional complexity at the Candida-host interface is that C. albicans has several developmental cell types including yeast, pseudohyphae, and hyphae, each with different modes of interaction with innate immune cells [15,16]. The ability of C. albicans to switch between the yeast and filamentous forms is strongly associated with virulence. Filaments are distinct from yeast-form cells in cell wall structure, cell wall proteins and transcriptional programs. In addition, the hyphal form of the fungus has been shown to cause more tissue damage than the yeast-form fungus in ex vivo models of candidiasis [17,18,19]. Furthermore, immune recognition of yeast provokes a different immune response compared to recognition of hyphae [15,20]. Intriguing recent work has suggested that only yeast-form cells possess exposed b-glucan at a few sites [9], but this has yet to be shown in vivo. Although b-glucan can be an important signal for innate immune cells in vitro, and the b-glucan receptor Dectin-1 protects against fungal disease in mice, we still understand little of the dynamics of b-glucan-Dectin-1 interaction during infection or drug treatment. We have developed a new technique to directly measure morphotype-specific b-glucan exposure in mouse tissue Candida is a common human commensal but disseminated candidiasis is a serious clinical problem, especially among immunocompromised patients. The innate immune system controls Candida infection, in part through the germline-encoded b-glucan receptor Dectin-1. However, during in vitro growth, Candida albicans mutes Dectin-1 recognition by cloaking its b-glucan underneath a layer of mannan. Bridging these two seemingly contradictory observations, we demonstrate that C. albicans masks b-glucan early during infection, but it becomes exposed later, allowing Dectin-1 to recognize the fungi and mediate immunity. Remarkably, treatment of mice with subtherapeutic doses of the antifungal drug caspofungin causes exposure of b-glucan on C. albicans even when it would not be exposed naturally. We introduce a new technique for monitoring of epitope exposure during infection, which can be used to monitor the availability of any epitope for immune recognition. This technique allowed us to show that natural unmasking of b-glucan is not morphotype-specific, but drug-mediated unmasking is biased towards the invasive filamentous form of C. albicans. These results highlight the unexplored area of dynamic epitope exposure during infection and therapy, which might be targetable to enhance immune recognition and fungal clearance. during infection and anti-fungal treatment. We find unexpected dynamics of b-glucan exposure during infection and du (...truncated)